Maybe the board was shortcutted to the box and got fried a tiny bit.
It was elevated from the box by unconventional, a bit broader (compared to what is screwed to it now) spacers im the beginning. Especially the upper left corner looks like the first soldering is very close to the screw hole. Maybe the board got short cutted there.
The question remains whether the board is “fixed” now or will show repeated failures in the future. I will reconnect everything in the next days and observe it’s behaviour. If I decide to get a new board, I will install it with plastic washers between the board and the correct slim spacers.
Hi Aaron,
I didn’t see that thread, many thanks! Will try that, especially because the connection seemed very weak to me before.
But this is only for the black TFT cable, correct? I am wondering why that should interfere with the marlin mode, which uses the grey ribbon cables if I am not mistaken.
That back side view looks to me like the solder work on those driver headers is pretty suspect.
If you have it out again, any chance of a close-up on at least one driver set?
I suspect that you have intermittent solder joints from bad workmanship and this will be a problem at times until those get re-flowed.
I agree with MakerJim. The reason the solder looks suspect to me… the top most slot appears to have what may be a cold solder joint (grey looking with dark ring around pin)… yet the others that melted appear they were reflowed (which would happen with all the heat). Maybe it reflowed itself enough to work again?
Wow. I really know nothing about boards and soldering thanks for staying with me and your suggestion
I am not 100% sure whether you wanted a close up of the back side of one of the driver slots, but that would be here in different angles, as good as my camera can do.
Well, it would be awesome if it somehow fixed itself. When it did, it should be permanent, correct? Meaning, if I pull and re-insert the drivers, they should work all the time. That would be easy to find out and I highly doubt it
About the scrambled TFT, I read others posts on Reddit, describing the same problem also with other boards. With some describing, that they have the exact same configuration on several machines and only some exhibit these problems. Which would indicate a not to rare hardware problem of the BTT TFT35 V3 E3. Is there anything that I can only do with the TFT in marlin mode? Otherwise I’d ignore this problem until I stumble over a solution and use the touch mode and PC for everything.
Yeah relying on “self reflowed” connections is a bit sketchy. If they do work reliably… ie replug all of them several times and no fails… then it may be reasonable to look the other way. Personally, I’d resolder new connectors on… but I do crazy things all the time. Worst case though, is a driver disconnecting in the middle of an expensive peice of stock that has had several hours of work put into it already. Replacement hardware isn’t cheap, but the alternatives that come with sketchy hardware can sometimes be worse.
I personally only use Marlin mode on my printer… don’t use a tft on my cnc, since it has an rpi with a gcode sender on it. AFAIK, you don’t strictly need touch mode to use a stock mpcnc. I have read some threads where folks are working on getting touch mode to be more cnc friendly. Having a working touch mode may be a good thing as that matures.
Replying again after seeing something in the latest photo post… the board I think should be replaced. There are several areas on the underside of the PCB that appear to have heat damaged solder resists (areas where the “paint” looks crinkled up… may flake right off with a light touch). If this is the case, the PCB has been compromised too much for continued use IMHO.
Next board, definitely do take care of insuring those standoffs aren’t shorting traces. I bought a kit of nylon standoffs from Amazon that I use for this sort of thing if I don’t feel like printing a mount. The kit came with assorted length nylon standoffs, nylon nuts, and nylon screws. So you never have to have metal touching a PCB.
I’m in agreement with Kev- I wouldn’t trust that board after whatever happened to cook it.
If I were really in a bind, I might scrape off any charring and try to white wire any stressed traces- but if I were going to do something like that, it’d mainly be to use it as a software development mule and not as a production board.
Probably the most important thing for you to do now is to figure out if that overcurrent and meltdown happened because something in your setup is allowing the through hole leads to short on something (a case or a metal plate, etc.) If so, you should fix that ASAP.
Sad to say, but those drivers are suspect too- if something in that system got hot enough to do that, they are likely to have been stressed.
Is there any chance you got a “refurbished” board. I’ve seen crap like that on quite a few things I got from Amazon- due I suspect to their easy return policy and less than scrupulous sellers.
I’ll get a new board and will try the current drivers first. If they act suspecious, I’ll order new ones.
If I understood you correctly, the problem is a combination of a bad quality board and (most likely) a short in my setup which finally cooked the board. Would you say the board quality is a warranty case? Well, I got it from the v1 shop so Jeff and Ryan can follow the case thereselves
Let’s see, maybe the new board will also improve the TFT behaviour…
I doubt that the issues that I described about Amazon returned products and board quality apply to the boards that Ryan sells. I have a board and TFT that I purchased when I started working on a Repeat V3 build- and it was flawless when I looked it over on receipt.
It will be interesting to see what happens with your next board.
To follow up on an item from above- is there anything in your setup where those through hole leads can short to a case or other metal item? That’s probably the most important aspect that’s still a little unclear to me.
I’d feel a lot better about your plans to improve the situation. Isolation tape is a good idea. I like using Kapton for that, but any insulative tape that you can cover anyplace where a conducive surface is close to circuits- particularly through hole leads- is a good idea.
That impovised spacer looks like a thread-in wood insert, and it looks like a trace is directly under that. That’s never a good plan.
Yes like the person above me said some isolation tape from staples should be a good idea. Also a easy stand off could be as simple as a piece of plastic or rubber tubing cut to same lengths.
Did you end up getting a new board ?
Hi guys,
The new board arrived, got flashed, got tested, build in and works perfectly.
The drivers just slided in with no resistance at all. Couldn’t get them in completely on the last board no matter what.
Also the TFT scrambling is much better. Not completely gone but huge improvement.
Many many thanks to everyone involved in the solution finding process for my initial problem! It was a delight to work with you and I learned a lot
When comparing the old and new board, I noticed, that two chips (?) have different positions; the ones next to the jumper that defines power input. Seems to play no role?
Edit: I have one question that might be connected to the previous issue. When the stepper motors moved once and are then in resting mode, they buzz. I read that “low quality drivers” can be the reason for that. All my motors do the buzz. Could it be, that my driver’s were compromised when the board got fried? Or is the buzz unavoidable?
I still have some questions but these will need another thread if I can’t find the answers anyways.
thanks for staying with me and your suggestion
